Although renal transplantation has proved to be both useful and cost- effective compared to dialysis, complications are frequent. Much of the success of renal transplantation is dependent upon prompt identification of the cause of impairment of function so that appropriate treatment can be applied before the condition becomes irreversible. Though nuclear medicine systems and several methods of analysis are available to analyze and evaluate kidney function, there is currently no analysis system that combines information from all of the time-activity curve in providing a global assessment of renal function. An improved means of radionuclide renogram analysis would contribute to earlier diagnosis, improved accuracy of diagnosis, and reduced need for biopsies. An automated program would save technologists time and would make this more detailed analysis practicable. Presenting results as probability of diagnosis in each category would be very helpful to the clinicians and would make the degree of assurance of the diagnosis clear. From this sort of presentation good criteria for doing biopsies should develop. We propose to optimize the discriminating power of the model developed in Phase I, and incorporate it into an expert system, designed to recognize the renal pathophysiology and to support the radiologist and attending physician in their efforts to manage the transplant patient. The added discriminating power provided by the physiologic model will be evaluated. Software will be developed in such a way that it should be applicable to the usual nuclear medicine computers or to work stations linked to them. Thus, if successful, the product should be widely applicable to any facility with a renal transplant program desiring analytical treatment of radionuclide transplant evaluations.